A series of compounds which are either homo or heterodimers of specific dideoxynucleosides have been synthesized. In various assays the phosphate- linked dimers of such antiviral monomeric dideoxynucleosides possess greater activity than the combined activity of the dideoxynucleosides administered alone. Preliminary pharmacokinetic data obtained for one of the agents after intravenous administration to rats indicated that it crossed the blood brain barrier with a distribution that compared favorably to AZT. We propose to expand the series of dimeric compounds by synthesizing drugs which include the following nucleosides: ddA, ddI, ddc, ddG, and AZT. Since the lipophilicity and therefore the CNS bioavailability of the specific agents appears to be greater for the esterified analog of the nucleoside dimer triester we also intend to prepare a series of triesters to examine the impact of such substitution on anti-viral activity and cytotoxicity. Based on the in vitro results that are obtained, we will examine the in vivo distribution and metabolism of the most active compounds. The sum of the information we obtain from these experiments will allow us to chose a specific compound(s) which will be pursued in Phase II studies.